Per. Mineral. (2002), 71, SPECIAL ISSUE: Archaeometry and Cultural Heritage, 101-111
PERIODICO di MINERALOGIA established in 1930
http://go.to/permin
An International Journal of MINERALOGY, CRYSTALLOGRAPHY, GEOCHEMISTRY, ORE DEPOSITS, PETROLOGY, VOLCANOLOGY and applied topics on Environment, Archaeometry and Cultural Heritage
Petro-archaeometric characterisation of "cotto ferrarese": bricks and terracotta elements from historic buildings of Ferrara GIANLUCA BIANCHINI*, ANNALISA MARTUCCI and CARMELA VACCARO
Dipartimento di Scienze della Terra - Universita degli studi di Ferrara - Corso Ercole I D'Este 32, 1-44100 Ferrara, Italy
ABSTRACT. Petro-archaeometric characterisation of «cotto ferrarese», i.e., chemicalmineralogical investigation of bricks, tiles and decorative elements from historic buildings in Ferrara (NE Italy) enabled us to constrain the nature of the raw materials employed. These consist of clays closely resembling the present-day fine sediments of the river Po (High-Cr sediments); clays of plausible Apennine provenance (Lo w-Cr sediments), although outcropping in the area, were not used. Processing of chemical data reveals that, within the High-Cr clays, the most CaO-rich facies were preferred and mixed with minor amounts of the local sands; a Na-Ovrich flux component was then probably added. After firing tests on these sediments, the related firing parageneses were compared to those recorded in the studied terracottas, indicating for the latter kiln temperatures between 850-1000°C, with significant variations among samples from different buildings. RIASSUNTO. - In occasione di recenti interventi di restauro e stato. possibile studiare dal punto di vista petro-archeometrico i laterizi di alcuni edifici storici del centro di Ferrara. In tale contesto, sono stati investigati mattoni ed elementi in cotto dei seguenti monumenti: Chiesa di Santa Maria in Vado (fondata nel X sec. e quasi totalmente riedificata nel XV-XVI sec.), Monastero Sant' Antonio in Polesine (XII-XVI sec.), Chiesa di Santo Stefano (fondata nel X sec. e riedificata nel XV-XVI sec.), Cattedrale di Ferrara (abside; XV -XVI sec.), Palazzo Schifanoia (Sala degli Stucchi; XV-XVI sec.).
* Corresponding author, E-mail:
[email protected]
La realizzazione di una banca-dati contenente analisi chimiche (XRF) ed analisi diffrattometriche (XRD) e risultata utile per interpretare i relativi processi produttivi nonche la natura dei materiali impiegati. I dati composizionali dei manufatti oggetto di studio, confrontati con i dati geochimicomineralogici dei sedimenti affioranti nell' areale ferrarese, hanno confermato una provenienza locale delle materie prime, suggerendo un uso esclusivo di sedimenti argillosi legati ad eventi alluvionali del fiume Po (tipologia: High-Cr); fra questi si e riscontrato I' utilizzo preferenziale delle facies particolarmente ricche in CaO. L'uso di sedimenti ad affinita appenninica (tipologia: Low-Cr), benche affioranti nell' areale ferrarese, non e stata invece riscontrata in detti laterizi. Sistematiche differenze nelle concentrazioni di alcuni elementi maggiori (es.: Si0 2, A120 3 , CaO, Na20, K 20) fra i laterizi investigati e le materie prime argillose, non imputabili a deidratazione/decarbonatazione durante la cottura, sono plausibilmente dovute a correzioni dell' originario impasto argilloso con sabbia quarzosa-feldspatica di provenienza locale, come indicato dai «trends» chimici e dai caratteri tessiturali-strutturali osservabili nell' analisi petrografica. In alcuni campioni e stato inoltre osservato un anomalo contenuto in sodio che non puo essere spiegato con I' aggiunta di un degrassante di natura quarzosa-feldspatica, rna potrebbe essere imputabile all'uso di un fondente ricco in sodio. Campioni rappresentativi delle argille locali sono stati sottoposti a prove di cottura a temperature di 750, 800, 850, 900, 950, 1000 °C per verificare l'evoluzione della paragenesi mineralogica al variare delle condizioni termiche, e per interpretare poi le condizioni di cottura dei menzionati laterizi storici;
102
G.
BIANCHINI,
A.
MARTUCCI
sana state cost riscontrate temperature di cottura fra i 850-1000 °C, can significative differenze fra i laterizi appartenenti ai diversi edifici storici oggetto di studio.
INTRODUCTION
Geology (i.e., outcropping lithotypes) influences raw material availability, and thus building methods. In Ferrara (NE Italy), situated in the Po alluvial plain, the widespread presence of silico-clastic sediments (and the lack of any rock outcrops) led to the use of bricks (X x + +;.t + "x + 0 0 + +
g ~OOCO
15
0
;1(* ;1(;1( ;I(
lO
;I(
Si02 5 50 16
60
55
CaO
14
+
12
~
•
+
;1(;1(
~CO
10
~
;XXO x ~oo x+X+0 + + cD 0
+
80
75
70
65
;1(;1(
;I(
+
;I(
+ + + +
Si02
0 50 4,5
65
60
55
80
75
70
K20
4,0
+ +
XXX 3,5
X X
+
\~~
3,0 2,5
0 Q+ 00 O:l(~
+
o
2,0
0
:l(
0
;I( ;I(
:l(
Si02
1,5 50
60
55
65
75
70
80
3,0
Na20
...
•
2,5
x
0
2,0
:l(:l(
"~CD 180 ppm; Ni>100 ppm) and low (Cr 900-950 DC), probably due to progressive destabilisation of other CaO-bearing phases such as melilite and amphibole; - plagioclase predominates over alkaline feldspar in the same thermal conditions of clinopyroxene formation (>950 DC), probably due to progressive destabilisation of both melilite and illite; - loss of cristallinity at increasing firing temperature was evaluated from the steadily increasing back ground intensity of the XRD diffraction patterns. These results are summarised in Fig. 4, which highlights that the terracottas from Sant' Antonio in Pole sine and Santa Maria in Vado were fired at temperatures close to 850 DC, whereas those of Santo Stefano were fired at significantly higher temperatures (up to 1000 DC). Instead, samples from the Cathedral of Ferrara show a wide range of firing temperatures, probably indicating heterogeneous thermal conditions (different furnace kilns?) during their preparation. For more reliable information, further starting compositions should be investigated in firing tests with monitored redox conditions. Moreover, considering the possible correction of the original clay bodies with sand (added to
108
G.
BIANCHINI,
A.
MARTUCCI
and C. VACCARO
Sant Ant onio in Pole sine Monastery I
.-----------"----~
Santa Maria in Vado Church
750
800
Santo
850
Stefano
Church
1000
Firing Temperature Fig. 4 - Thermal constraints provided by firing tests (integrated with available literature data): stability fields of mineral phases recognised within the terracotta parageneses. Firing temperatures estimated in bricks of different historic buildings are reported, to identify the relative technological processes. See text for further details.
109
Petro-archaeometric characterisation of "cotto ferrarese" ...
limit shrinkage) and/or with small amounts of a Na-rich flux (which may displace firing reactions towards lower temperatures), it would be interesting to simulate the effect of adding temper (and/or flux).
CONCLUSION
The presented data attest that the terracottas of the historic buildings of Ferrara were prepared with the CaO-richest clays occurring around Ferrara at that time (High-Cr clays). As shown in Fig. 5, these predominant materials were often corrected by adding local sands. Instead, the further addition of a Na-rich flux is suggested by the Na20 excess (relative to the local sediment composition) of the studied materials.
Compositions are reported in the Fe203tot(Na20+K20)-(CaO+MgO) ternary diagram (Fig. 6a), proposed by Fiori et al. (1989) to classify ceramic raw materials, and plot on the Cottoforte and Maiolica compositional fields. Further characterisation of these materials is provided by the (CaO+MgO)-AI 20 3-Si02 diagram (Fig. 6b), which shows the composition of phases usually observed in bricks, as well as the tie-lines dividing the relative stability fields (Artioli et al., 2000); our samples and the local clay composition mainly plot in the subtriangle Si- Wo(Di)-An, indicating that quartz - wollastonite (diopside) plagioclase is the predicted stable paragenesis for terracottas of similar bulk composition. In this light, the presence of carbonates and/or melilite (recorded in several Ferrara bricks and «cotto» elements) indicates that equilibrium
2.0 ......- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ,
Herron, 1988: Chemical c1assifi cat ion of terrigenous sediment s
Log( Fe203/ K20) 1.5
\
Fe-shale 1.0 -Bricks (t iles and cotto
decorat ive elements) of historical Ferrara buildings
Fe-Sand
0.5
shale
Sub lit harenit e
0.0
Oays outcropping in t he Ferrara surrounding fifO
Subarkose
~
-0.5
Po sands Ferrara
0.2
0.4
0.6
0.8
1.0
1.2
Log (Si02/AI 203) 1.4
1.6
1.8
Fig. 5 - Bricks and local clay composition plotted in Log (Fe203/K20) vs. Log (Si0 2/AI20 3) diagram reporting compositional fields for various terrigenous sediments (Herron, 1988).
110
G. BIANCHINI, A. MARTUCCI and C. VACCARO
CaO+MgO
a) Bricks (tiles and cotto decorative elemen ts) of Ferrara his torical buildings.
Ferrara
Fe203
Na20+K20
Si02 Bricks (tiles and cott 0 decorat ive elemen ts) of Ferrara his tori cal buildings.
b)
CaO+MgO
AI20S
Fig. 6 - a) Ferrara historical terracottas and local clay compositions plotted in Fe203-(Na20+K20)-(CaO+MgO) diagram reporting geochemical characterisation for ceramic raw materials (Fiori et al., 1989); b) Ferrara historical terracottas and local clay compositions plotted in (CaO+MgO)-AI 20 rSi0 2 diagram reporting composition of phases usually observed in firing tests (with relative tie-lines; Artioli et al., 2000).
Petro-archaeometric characterisation of "cotta [errarese" ...
conditions were not attained. This is ascribed to the lack of homogeneity of the original starting material, which consists of several microdomains of different composition, and indicates limited mobility of the chemical species during firing (Duminuco et al., 1996, 1998; Riccardi
et al., 1999). For this reason, a detailed investigation of the reactions occurring during firing processes (and thus a better calibration of the estimated firing temperatures) would require integration of the reported «bulk» data with in-situ analyses.
ACKNOWLEDGMENTS Profs. L. Beccaluva and F. Siena are kindly acknowledged as the promoters of this research and for their helpful discussion. The authors are also grateful to Prof. P. Brotzu and Dr. M. Dondi for the useful suggestions reported in their constructive reviews.
REFERENCES ARTIOLl G., BAGNASCO GIANNI G., BRUNI S., CARIATI F., FERMO P., MORIN S. and Russo U. (2000) - Studio spettroscopico della tecnologia di cottura di ceramiche etrusche dagli scavi di Tarquinia. In Atti del I Congresso di Archeometria (Patron Editore- Bologna), 335-349. BIANCHINI G., LAVIANO R., Lovo S. and VACCARO C. (2000) Clay sediments in the area surrounding Ferrara (Italy): a contribution for the environmental monitoring. Plinius (Italian supplement of the Eur. J. Mineral.), 24, 38-39. BIANCHINI G., LAVIANO R., Lovo S. and VACCARO C. (2002) Chemical-Mineralogical characterisation of clay sediments around Ferrara (Italy) a tool for an environmental analysis. Forthcoming publication. BONDESAN M. (1990) - Evoluzione geomorfologica ed idrog rafica della pianura ferrarese. In G. Corbo Ed., «Terre ed acqua-Le bonifiche ferraresi nel delta del Po», 13-20. BRINDLEY G. and LEMAITRE 1. (1987) - Thermal,
111
oxidation and reduction reactions of clay minerals. Mineral. Soc. Monograph. 6, A.C.D., Newman ed. CAPEL J., HUERTAS F. and LINARES J., (1985) High temperatures reactions and use of Bronze Age pottery from La Mancha, Central Spain. Miner. Petr. Acta, 29-A, 563-575. CHAPMAN V.l. and CHAPMAN D.S. (1980) Seaweeds and their uses. Chapman and Hall. CULTRONE G., SEBASTIAN-PARDO E., CAZALLA 0., RODRIGUEZ-NAVARRO C. and DE LA TORRE M.J. 2000 - Mineralogical changes during brick production in laboratory experiments. In QuarryLaboratory -Monumen t International congress, Pavia 2000 (proceeding volume n.1), 253-258. DUMINUCO P., RICCARDI M.P., MESSIGA B. and SETTI M. (1996) Mo dific azioni tessiturali e mineralogiche come indicatori della dinamica del processo di cottura dei manufatti ceramici, Ceramurgia, XXXVI, 281-288. DUMINUCO P., MESSIGA B. and RICCARDI M.P. (1998) - Firing processes of natural clays. Some microtextures and related phase compositions. Thermochimica acta, 321, 185-190. FIORI C., FABBRI B. and RAVAGLIOLI A. (1989) Maurie prime c eramich.e vol. I. Biblioteca Tecnica Ceramica, Faenza ed. FRANZINI M., LEONI L. and SAITTA M. (1975) Revisione di una metodologia analitica per fluorescenza-X basata sulla correzione completa degli effeui di matrice. Rend Soc. It. Min. Petrol., 31, 365-378. HERRON M.M. (1988) - Geochemical classification of terrigenous sands and shales from core or log data. J. Sed. Petrol., 58,820-829. LEONI L. and SAITTA M. (1975) X-ray fluorescence analysis of 29 trace elements in rocks and mineral standards. Rend Soc. It. Min. Petrol., 32,497-510. MARSIGLI M. and DONDI M. (1997) - Plasticita delle argille italiane per laterizi e previsione del loro comportamento in forgiatura. L'industria dei 1aterizi, 46, 214-222. RICCARDI M.P., MESSIGA B. and DUMINUCO P. (1999) - An approach to the dynamics of clay firing. Applied Clay Science, 15,393-409 ROLLINSON H. (1993) - Using geochemical data: evaluation, presentation, interpretation. Longman Publishers Ltd. STIAFFINI D. (1999) - II vetro nel Medioevo. Fratelli Palombi Eds.
